Research On Train Control Method Of Virtual Coupling Oriented To Dynamic Coupling And Close Tracking

To address the ever-increasing need for capacity in the rail transit industry,the concept of virtual coupling arises at the historic moment,and gradually becomes one of the research hotspots in the field of train control.Virtual coupling uses wireless communication instead of mechanical couplet,and the train tracking interval will be greatly shorten.Although this mode of train operation improves the efficiency of transportation,the dynamic coupling and the extremely low tracking interval after dynamic coupling bring great challenges to train control.At present,there is no train control strategy suitable for virtual coupling at home and abroad.Therefore,this thesis researches the dynamic coupling process and close tracking process after dynamic coupling,and proposes corresponding control methods.The main work of this thesis is as follows:(1)Based on the functional requirements of virtual coupling,the composition of train control system suitable for virtual coupling is discussed,and the train control principle in dynamic coupling and close tracking process is expatiated based on the system.Then the calculation method of minimum tracking interval of coupling train is proposed,which lays a foundation for the proposal of virtual coupling control method.(2)In order to realize dynamic coupling in train operation,an optimal control method for dynamic coupling is proposed,which combines offline speed curve planning with online optimization control.Firstly,the optimization problem is constructed with the objective of accurate coupling and low energy consumption,and the dynamic programming algorithm is used to solve the problem,and the offline optimal speed curve is obtained.Then,in the actual operation state,the optimization problem is established according to the real-time interval in a rolling manner,the offline speed result is taken as the initial solution,and the sequential quadratic programming algorithm is used to solve the optimal control strategy online,which ensures the high computational efficiency.(3)An adaptive control method based on radial basis function neural network is proposed to realize the close tracking of adjacent trains in the coupling convoy.In this method,the speed of the leading train is adjusted as the reference speed.Then the neural network model whose output is the control force is constructed,and the network is trained with the target of reducing the difference between the actual speed and the reference speed.In order to achieve the control objective of stabling tracking interval near the ideal interval,the network training strategy needs to be improved.The target of network training is dynamically adjusted according to the real-time tracking interval,and the output of the trained network is used to control the train.(4)Simulation experiments of virtual coupling train control are carried out,and the control methods proposed in this thesis are applied to virtual coupling train.The experimental results show that the proposed dynamic coupling control method can achieve the objective of accurate coupling and low energy consumption under the condition of external disturbance,and has higher computational efficiency than the ordinary optimization method.The proposed close tracking adaptive control method can synchronize the speed of adjacent coupling trains,and has the ability of adaptive adjustment of tracking interval.The proposed train control methods suitable for virtual coupling can realize the safe operation of the train in dynamic coupling process and close tracking process,which has certain reference significance for the follow-up research of virtual coupling.This thesis includes 56 figures,12 tables,63 references.